A Comparison of Methods to Predict Soil Surface Texture in an Alluvial Basin

Abstract

Surface soil texture controls many important ecological, hydrological, and geomorphic processes in arid regions and is therefore important from a land-management perspective. Soil survey efforts have traditionally fulfilled this need, but they are constrained by the size, remoteness, and inaccessibility of many arid regions, which renders simple field measurements prohibitively expensive. This article compares several different predictive soil-mapping techniques with a sparse data set in order to develop surficial soil texture maps. Our results suggest that data collected at the landscape scale can be used as input to predictive soil-mapping techniques to create maps of soil texture at higher fidelity and a fraction of the cost than would be required using traditional methods.

Key Words:

• predictive soil mapping
• soil texture
• soil survey
• geostatistics
• Joshua Tree National Park

Notes

¹The names used to refer to the variables in the text are in parantheses.

²TM wavelength ranges: Band 1=0.45–0.52 μm; Band 2=0.52–0.60 μm; Band 3=0.63–0.69 μm; Band 4=0.76–0.90 μm; Band 5=1.55–1.75 μm; Band 7=2.08–2.35 μm.

³NDVI=[(TM4–TM3)/(TM4+TM3)]

¹Prob. values from difference in means t-test between the best performing model (MLR) and the others.

¹Prob. values from difference in means t-test between the best performing model (MLR) and the others.

Additional information

Notes on contributors

Peter Scull

An Assistant Professor

Greg Okin

An Assistant Professor

Oliver A. Chadwick

A Professor of Environmental Studies and Geography

Janet Franklin

A Professor of Biology and Adjunct Professor of Geography